Research: (under constant change)

I am interested in understanding the processes and mechanisms involved in learning and in how memories are formed. My research and interests are guided by questions such as: How do we learn? What changes occur in the brain when we learn? What chemicals and processes are critical for producing those changes?

Over the years I have used a number of techniques to investigate issues related to the central question of how we as humans learn. The techniques I've used include: histology, electrophysiology, pharmacology and behavior.

Most recently, experiments have been conducted with assistance of undergraduate students at Oberlin College addressing the role of certain proteins (voltage dependent calcium channels) in the formation of long-term memories. Some of these students have since moved on to research laboratories or graduate programs. For example, my first Honor's student, Alex Goddard, is currently working on his Ph.D. in Neuroscience at Harvard, Bonnie Fletcher is at NYU. More recent graduates are working in New Zealand at the University of Aukland (Wes Clapp), and at Boston University (Lara Petrak).

The rest of this page highlights some of the work done by Oberlin students in my lab. Even though some never made it to publication, much of it was very illuminating within the context of the question being asked.

some photos taken by Gisela Knight

[ Home ] [Classes ] [Travel] [Research] [OCTET]

Kunal Dalal ('01), Bonnie Fletcher ('00) & Lara Petrak ('01)

This project asked the question "Would the presence of voltage dependent calcium channel antagonists prevent the induction of LTP induced in vivo (i.e. in an awake moving animal)." The question arose from a report in which researchers at other laboratories had shown that there seem to be two forms of synaptic modifications that can occur. The first requires the activation of NMDA receptors and is short lasting. The second requires activation of voltage dependent calcium channels and is longer lasting. These results were demonstrated in vitro and in anesthetized animals. However, in vitro preparations and anesthetized animals are not necessarily do not reflect the state of neuronal tissue as it normally is when it is learning! This project was incredibly difficult. 4 electrodes needed to be implanted. A recording electrode was placed in the CA1 region of the right hippocampus. Two stimulating electrodes were positioned in the CA3 region, one on the right and one on the left side. The fourth electord was a ground wire placed around the nearest screw. Electordes and the head assembly were then cemented into place. Subjects were run 2 at a time and changes in evoked responses were monitored over a 10 day period. Results were mixed and did not allow for any conclusion to be reached. However, much technical information was gathered. Therefore the question of whether or not VDCC LTP is evident in an awake moving animal is still open.

More Pictures

Click on image to see larger version	Short rationale behind this project.  Activity dependent calcium increases in neurons is important for changing  the strength of connections between neurons.  This process is thought to be important for the formation of memory.  While it is known that the blockade of NMDA channels (one means for increasing intracellular calcium)  can prevent  rats from learning the nuances of  an 8-arm radial maze, it is not known what effect the blockade of voltage dependent calcium channels (VDCCs - another means of increasing intracellular calcium) will have on a similar task. This project was designed to discover if calcium entry via VDCCs might also be involved in the learning of the 8-arm radial maze.  More details about the background and  procedure  are described elsewhere.     During this winter term  2 8-arm radial  mazes located  in the Sperry building were continually in use from 11:00am - 8:30pm.  This was made possible by the diligent work habits of nineteen  students (see left margin) and the quick and beautiful work of Bill Martin in the machine shop.  Students were divided into 8 groups (2-3 students/group). Each group was  responsible for 4 rats. Students  injected, ran the rats daily (including 2 weekends) and scored the rat's progress.  Bonnie Fletcher and Harlan  Fichtenholtz  were assigned to deal with data analysis and students from my Neurobiology of Learning and Memory laboratory were in charge of overseeing certain timeslots.       The project was a huge success.  I've was very happy with the effort and care that the students have taken in being on time and being gentle with the animals. Also I would like to thank those that provided space and funding:  Neuroscience Dept., Winter Term Committee and the Dean's Office.  Further thands to Dr. Teyler who helped design this study and Gigi Knight. All procedures were approved by IACUC (#99RNAB1).  Preliminary Results !!      With respect to learning which arms had food in them and which didn't (reference memory -RM), the  group of animals given the VDCC blockers learned the task well, but failed to retain it.  The group of animals given the NMDA blocker didn't learn as well, but retained their learning over the retention interval...and even got better!  The combined drug group learned the worst of all, and didn't  retain their meager learning over the retention interval.     With respect to learning not to reenter an arm where they already found food (working memory -WM),  all groups did pretty well except for the combined drug group who showed a lot of WM errors.     With respect to performance on the first and second trials of each day,  there does not seem to be a clear pattern with respect to RM errors.   On WM errors, however, most drug groups show fewer errors on the 2nd trial. This reduction in reference errors is most evident in the group where both NMDA and VDCC channels were blocked.    Graphs - page is not optimized and therefore may take some time to load! Conclusion    Blocking VDCCs does affect learning.  Verapamil (VDCC antagonist) and MK801 (NMDA antagonist) differentially affect learning as assayed by comparing reference errors and  the ability of the animals to retain the information over a period of 1 week.      The combined drug seems to have an effect on learning that is more than the sum of the individual drugs alone.      Further studies that have been suggested: Testing the ability of these drugs to prevent the induction of synaptic plasticity in awake animals. Testing the effect of these drugs on the learning of a new food pattern. Testing the effect of these drugs on visual discrimination.

 
 

Transluminated.	Illuminated from above. Cuts were made to determine the trajectory of the fibers responsible for evoked response.
Graph illustrates the result of a sustained 1 Hz tetanus on evoked responses recorded at different electrodes in the contralateral hemisphere. Pretetanus recordings are traced with a solid line. Post tetanus recordings are traced with broken lines. The different colors represent the different electrodes (1, 5, 10 & 15).

The methods for recording and analysizing the data are described in
METHODS FOR STUDYING THE CONDUCTANCE CHANGES ASSOCIATED WITH SYNAPTIC ACTIVATION OF FOREBRAIN SLICES - THE INTERPRETATION OF FIELD POTENTIALS USING CSD PROFILES
BORRONI AM, VAKNIN G, BERRY R, TEYLER TJ
JOURNAL OF NEUROSCIENCE METHODS
39 (1): 89-102 AUG 1991
&
AN INTEGRATED MULTIELECTRODE ELECTROPHYSIOLOGY SYSTEM
BORRONI A, CHEN FM, LECURSI N, GROVER LM, TEYLER TJ
JOURNAL OF NEUROSCIENCE METHODS
36 (2-3): 177-184 FEB 1991
___________________________________

The methods described above were used to investigate other phenomenon as described in
HYPERPOLARIZING AND DEPOLARIZING GABA-A RECEPTOR-MEDIATED DENDRITIC INHIBITION IN AREA CA1 OF THE RAT HIPPOCAMPUS
LAMBERT NA, BORRONI AM, GROVER LM, TEYLER TJ
JOURNAL OF NEUROPHYSIOLOGY
66 (5): 1538-1548 NOV 1991

Equipment:

This work was an extension of the work described in:

This project was undertaken in collaboration with researchers at Louisiana State University, Tennessee State University, and the Northeastern Ohio Universities College of Medicine (Kent State University). This program is designed to mimic the ability of a rat to navigate in an open environment (spatial navigation).  The simulation consists of four layers of 'neurons' linked together in various ways.  The linkages and layer configurations are constrained by the anatomy of the rat hippocampal region - a region critical for spatial navigation.  Here at Oberlin we tested various network configurations and learning rules to see which combinations produced the best results  in terms of allowing our simulated rat to produce a find a 'goal' in it's open environment.   John Matney (OC'99) did most of the work. He collect pertinent data consisting of vector quantities calculated after the network had stabilized and the number of 'steps' the simulated rat needed to find the goal after it had time to orient itself to it's new environment.  John patiently collected data from 2016 different configurations. Along with the data collected by Matt Chow at NEOUCOM we were able to rank order the various configurations for efficacy in learning the task.  The next step will be to further legitimize the network with respect to hippocampal anatomy and determine which factors are most critical for learning to occur within a given network configuration.

We are running the simulator on a PowerPC.  The figure is an example of the interface. The box in the bottom right represents the field in which  the rat is navigating.  The red circle is the goal and the vectors are built during the explore phase and later used to find the goal during the search phase.

Sponsored independent study/private reading projects:

Ethan Meyers (Spring 2001) : Private reading: Neural Networks. Using Computational Neuroscience textbook by K & O'Reilly tried to create neural network usign C++ http://occs.cs.oberlin.edu/~emeyers/Research/research.html

[ Home ] [Classes ] [Travel] [Research] [OCTET]